AI Security Research Engineer: RL Training & Red Teaming

Qualis; Sunnyvale (CA); On-site; USD 120,000 - 180,000; Full time

Job summary

Qualis in Sunnyvale, CA seeks researchers to develop, fine-tune and deploy advanced AI models for autonomous agents in RL training environments. You will focus on red-teaming, post-training evaluation, and hardened security loops to prepare agents for real-world adversaries.

The role balances research and engineering with autonomy, allowing exploration of new techniques, staying current with AI/LLMs, and contributing to open-source AI projects. MS or PhD and ML experience are welcomed.

Qualifications

• MS or PhD in CS, AI, ML or similar from a top school.
• Experience in Pre or Post Training.
• Open to years of experience from junior-mid to senior level.
• Strong with ML frameworks and Python.
• Extra curricular applied research or open source contributions in AI are highly regarded.

Responsibilities

• Develop, fine tune and deploy advanced AI models.
• Implement fine-tuning techniques to adapt pre-trained LLMs to new tasks and domains.
• Research and stay up to date with AI, LLMs, model eval and fine-tuning techniques.
• Assist and contribute to adversarial LLM red-teaming, developing TTPs and using AI to find vulnerabilities.

Skills

Python

ML frameworks

Adversarial ML

Research experience

Education

MS/PhD in CS/AI/ML

Job description

Qualis in Sunnyvale, CA seeks researchers to develop, fine-tune and deploy advanced AI models for autonomous agents in RL training environments. You will focus on red-teaming, post-training evaluation, and hardened security loops to prepare agents for real-world adversaries.

The role balances research and engineering with autonomy, allowing exploration of new techniques, staying current with AI/LLMs, and contributing to open-source AI projects. MS or PhD and ML experience are welcomed.